diff options
| author | Erik Scholz <Green-Sky@users.noreply.github.com> | 2023-06-22 14:20:47 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2023-06-22 14:20:47 +0200 |
| commit | 7487137227eb32ed9b12156338b865cb29b2dfd1 (patch) | |
| tree | d65ed6238bf5a04519dc038114fe7cc332993720 /examples/train-text-from-scratch | |
| parent | bbca06e26949686d61a5126332680ba3cccf235c (diff) | |
rework convert.py to read hyper-parameters from config.json (#1958)
* Read hyper-parameters from HuggingFace-transformer config.json, if they exist, and fall back to guessing, like before otherwise.
This allows converting open_llama 3B and other non-standard model designs.
Diffstat (limited to 'examples/train-text-from-scratch')
0 files changed, 0 insertions, 0 deletions